テストモデル(・8・) / It is a TEST model

🎯 Overview / 槪要:

• An EXPERIMENTAL model that merges Illustrious and NoobAI eps1.1

  IllustriousとNoobAI eps1.1をマージした実験的なモデル

• Compatibility with LoRA trained on IllustriousXL v0.1

  IllustriousXL v0.1で学習したLoRAとの互換性

• Background, poses, and outfits are influenced by Illustrious

  背景、ポーズ、衣装などはIllustriousの影響が強い

• Face, art style, and effects are influenced by NoobAI eps1.1

  顔、画風、エフェクトなどはNoobAI eps1.1の影響が強い

• Most characters can be drawn without LoRA. When using LoRA, the character LoRA from Illustrious v0.1 and the style LoRA from NoobAI eps1.1 seem to be more suitable.

  LoRA無しでもほとんどのキャラクターは描ける。LoRAを使用する場合、キャラクターLoRAはIllustrious v0.1バージョン、画風LoRAはNoobAI eps1.1バージョンがより適している印象。

🎯 Prompt / プロンプト:

• Positive:

masterpiece, absurdres, very awa, best quality, highres, newest, anime coloring

• Negative:

worst quality, low quality, worst aesthetic, normal quality, bad quality, lowres, unfinished, old, bad anatomy, bad hands, amputee, long torso, missing limbs, long fingers, malformed fingers, missing fingers, extra fingers, 6 fingers, text, censorship, jpeg artifacts, scan artifacts, lossy-lossless, multiple views, 3d, abstract

🎯 Parameter / パラメーター:

• Sampling Method: Euler a (WebUI A1111) / Euler Ancestral CFG++ or Euler SMEA DY CFG++ (WebUI reForge)

• Schedule Type: SGM Uniform

• CFG Scale: 2~7

• Steps: 28